Olfactory marker protein regulates adipogenesis via the cAMP–IκBα pathway

Olfactory marker protein (OMP) regulates olfactory transduction and is also expressed in adipose tissue. Since it serves as a regulatory buffer for cyclic AMP (cAMP) levels, we hypothesized that it plays a role in modulating adipocyte differentiation. To determine the role of OMP in adipogenesis, we examined the differences in body weight, adipose tissue mass, and adipogenic or thermogenic gene expression between high-fat diet-fed control and Omp-knockout (KO) mice. cAMP production, adipogenic gene expression, and cAMP response element binding protein (CREB) phosphorylation were measured during the differentiation of 3T3-L1 preadipocytes and mouse embryonic fibroblasts (MEFs). RNA sequencing was performed to determine the gene expression patterns responsible for the reduction in adipogenesis when Omp was deleted. Body weight, adipose tissue mass, and adipocyte size decreased in Omp-KO mice. Furthermore, cAMP production and CREB phosphorylation reduced during adipogenesis induced in Omp-/- MEFs, and the Nuclear factor kappa B was activated due to significantly reduced expression of its inhibitor. Collectively, our results suggest that loss of OMP function inhibits adipogenesis by affecting adipocyte differentiation.


Introduction
Olfactory marker protein (OMP), a cytoplasmic low-molecularweight protein found in olfactory receptor (OR) cells, sensory neurons, olfactory bulbs, and non-olfactory tissues, is commonly recognized as an OR marker (Margolis, 1972(Margolis, , 1982Nakashima et al., 1985). According to recent whole-genome sequencing studies, ORs exhibit both olfactory and non-olfactory activity, and OMP expressed in non-olfactory tissues transmits signals from OR-related processes (Kang and Koo, 2012;Kang et al., 2015). OMP expression in non-olfactory tissues indicates the existence of potentially associated events in non-olfactory systems. OMP has been extensively studied in OR neurons (ORNs). Recently, Nakashima et al. reported that OMP binds directly to cyclic AMP (cAMP) and, during the odorant response, appears to buffer fluctuations in ciliary cAMP to maintain ORN sensitivity (Nakashima et al., 2020). The results of this study prompted us to investigate whether OMP also affects non-olfactory tissues by modulating cAMP levels. In a previous study, conditional ablation of mature olfactory sensory neurones conferred better resistance to diet-induced obesity (Riera et al., 2017). However, whether OMP plays a physiological or functional role in adipogenesis has not yet been investigated.
cAMP, an important secondary messenger that helps in the proper functioning of sensory stimuli, hormones, and neurotransmitters, is positively correlated with the formation of adipocytes from preadipocytes (Yarwood, 2020). Preadipocytes develop into mature adipocytes upon stimulation with insulin, glucocorticoids, or other substances that increase intracellular cAMP levels. Elevated cellular cAMP concentrations have also been correlated with the expression of several transcription factors critical for adipocyte differentiation (Petersen et al., 2008). Considering that OMP contributes to the regulation of the kinetics and termination of olfactory transduction, we hypothesized that OMP may have a similar role in regulating adipogenesis owing to its involvement in cAMP signaling. Adipogenic stimulation induces DNA replication and cell cycle re-entry (mitotic clonal growth) through a transcription factor cascade, resulting in the expression of adipogenesis-related genes (Tang and Lane, 2012). Adipocyte differentiation, the main process of adipogenesis, is influenced by several transcription factors including CCAAT/enhancer-binding protein alpha and beta (C/EBPα and C/EBPβ), peroxisome proliferator-activated receptor gamma (PPARγ), and sterol regulatory element binding protein-1c . Upstream of C/EBP is the cAMP response element-binding protein (CREB) (Choi et al., 2014;Lefterova et al., 2014), a crucial interconnecting factor in cAMP signaling during adipocyte differentiation. cAMP selectively modulates nuclear factor kappa B (NF-κB), which plays a crucial role in switching gene expression and is a master regulator of adipocyte differentiation. The adenylyl cyclase (AC) activator forskolin and prostaglandin E2 inhibit NF-κB through an increase in the levels of its intracellular inhibitor (IκBα), a response mediated by cAMP (Neumann et al., 1995). A study utilizing NF-κB p65 fat-specific transgenic mice revealed that adipose tissue-specific overexpression of NF-κB p65 prevented preadipocyte differentiation into mature adipocytes, mostly via PPAR inhibition (Tang et al., 2010). These studies suggest that OMP alters adipocyte differentiation in the adipose tissue by altering cAMP levels. Furthermore, mechanisms related to its function may affect the activation of the NF-κB signaling pathway.
In this study, we investigated the role of OMP, both in vitro and in vivo, using adipogenic models and examined the mechanisms underlying the modulation of adipocyte differentiation to provide a broader understanding of the process.

Ethical statement
All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of Yonsei University Health System (approval number: 2020-0140).

Animal experiments
Omp − /− mice were originally developed by the Jackson Laboratory (Bar Harbor, ME, USA) and provided by Prof. Jae Hyung Koo, New Biology Research Center, DGIST (Daegu, Republic of Korea). All experimental animals were obtained through embryo transfer at the Yonsei Biomedical Research Institute (Seoul, Republic of Korea) and maintained under controlled conditions (12-h light/dark cycle in a temperature at 23 ± 1 • C and humidity at 50 ± 10%) with ad libitum access to food and water.
For the experiments, 4-week-old male mice were divided into four groups: normal diet (ND) (LabDiet Pico 5053, USA)-fed wild type (WT) mice (WT-ND, n = 7), high-fat diet (HFD) (60 kcal% fat, D12492, Saeronbio, Korea)-fed WT mice (WT-HFD, n = 9), ND-fed OMP knockout (KO) mice (KO-ND, n = 10), and HFD-fed OMP KO mice (KO-HFD, n = 10) (Rajpal et al., 2015). Three to five mice were housed per cage from the time of weaning until the end of the experiment. To mitigate the potential stress induced by the laboratory environment, we incorporated nestlets into mouse cages to provide a more comfortable and naturalistic environment. The food-finding test was performed according to the protocol described by Machado et al. (2018). Food intake was measured by providing 50 g of food to each cage (containing individually housed mice), and leftover food was weighed at 0.5, 3, 6, 9, and 12 h (Ellacott et al., 2010). The total body mass of the mice was recorded at weekly intervals over a course of 13 weeks. To measure body composition, dual-energy X-ray absorptiometry (DEXA) (INALYZER, Medikors, Korea) was performed at week 13, and the results were analyzed using INALYZER v1.4.0.0 software. At the end of the 13-week study, the mice were euthanized by CO 2 inhalation, and epididymal white adipose tissue (eWAT) and brown adipose tissue (BAT) samples were collected for further analysis. Approximately two-thirds of tissue samples from each mouse were fixed in 10% formalin, and paraffin tissue blocks were prepared for hamatoxylin and eosin (H&E) staining. The remaining tissue samples were stored in RNAlater solution (Thermo Fisher Scientific, Waltham, MA, USA) at − 70 • C until RNA extraction for the assessment of adipogenesis-related gene expression.

Plasmid transfection
The OMP (NM_011010) Mouse Tagged ORF Clone (MR201298) was purchased from OriGene Technologies, Inc. (Rockville, MD, USA). Before transfection, 3T3-L1 cells were seeded at a density of 3 × 10 6 cells/60-mm dish for 24 h. Thereafter, the cells were transfected with the appropriate expression plasmids using Polyjet transfection reagent (SL100688; SignaGen Laboratories, Frederick, MD, USA) and cultured at 37 • C for 48 h. Subsequently, the cells were treated with differentiation medium or dimethyl sulfoxide (D2650, Sigma Aldrich, St. Louis, MO, USA) for an additional 30 min before lysis. Ten days after the induction of differentiation, cells were washed twice in phosphate-buffered saline (PBS: LB201-02, WELGENE, Gyeongsan-si, Gyeongsangbuk-do, Republic of Korea), fixed for 1 h in 10% formalin, and rinsed three times in PBS. Oil Red O (O0625, Sigma Aldrich, St. Louis, MO, USA) was used to stain cells for 20 min in a filtered working solution. Subsequently, the cells were washed once with PBS and twice with distilled water and then air-dried. Oil Red O was extracted using 0.4 mL isopropanol, and the total lipid content was quantified by measuring the absorbance at 520 nm using a microplate reader (Multiskan Sky, Thermo Fisher Scientific, Waltham, MA, USA).

RNA isolation, RT-PCR, and quantitative real-time PCR
Total RNA was isolated using TRIzol Reagent (Invitrogen, Waltham, MA, USA), and cDNA was synthesized using ReverTra Ace (TOFSQ-101; TOYOBO, Osaka, Japan) according to the manufacturer's instructions. The resulting cDNA was subjected to either conventional PCR using Ready-2x-Go [Taq] (PMD001L, Nano Helix, Daejeon, South Korea) or quantitative real-time PCR using Power SYBR Green PCR Master Mix (#4367659, Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. The primers used in this study are presented in Table 1.

Intracellular cAMP measurement
After treatment of 3T3-L1 cells or MEFs, the cells were lysed in 1x cell lysis buffer (#9803, Cell Signaling Technology) containing 1 mM PMSF dissolved in isopropanol. The protein concentration was determined using a BCA protein assay kit (Thermo Fisher Scientific), following which equal amounts of protein were loaded. Intracellular cAMP production was assessed using a cAMP assay kit (#4339; Cell Signaling Technology) according to the manufacturer's instructions.

Table 1
Oligodeoxynucleotide primers used in the real-time quantitative reverse transcription PCR of epididymal white and brown adipose tissue samples of male 4week-old Omp − /− mice. Total RNA was extracted from previously isolated cells using the RNeasy Mini Kit (#74104, QIAGEN Ltd., Germantown, MD, USA), following the manufacturer's protocol. RNA sequencing was performed on an Illumina platform using the TruSeq Stranded Total RNA LT Sample Prep Kit (Gold) at Macrogen (Seoul, Republic of Korea).

Statistical analyses
For statistical analyses, Pearson's correlations were used to examine the correlations between OMP expression and adipocyte differentiation markers expression in 3T3-L1 cells. GraphPad Prism 6.0.1 software (GraphPad Inc., La Jolla, CA, USA) was utilized to automatically calculate Pearson correlation coefficients and corresponding p-values. A value of R close to 1 indicates a positive correlation and p < 0.05 considered statistically significant. Two groups were compared using unpaired Student's t-tests, and four groups were evaluated using twoway analysis of variance, followed by Bonferroni's post-hoc tests for multiple comparisons using GraphPad Prism 6.0.1 software. Each experiment was performed at least three times. Data are expressed as mean ± standard error of mean (SEM). Statistical significance was set at p < 0.05. Differences are indicated by asterisks (*p < 0.05, **p < 0.01, ***p < 0.001) or hash symbols (#p < 0.05; ##p < 0.01; ###p < 0.001). All measurements were obtained from distinct samples.

Whole-body KO of OMP causes diet-induced obesity resistance
First, we examined whether OMP plays a physiological role in weight gain and increases adipose tissue depending on the type of diet. As shown in Fig. 1A, the body weights of mice in the four groups were indistinguishable at weaning. However, weight was significantly lower in OMP KO groups than in WT groups from week 3 of administration in both the ND and HFD groups (#p < 0.05, ###p < 0.001 vs. WT-HFD) and showed a more significant difference in a time-dependent manner.
In the HFD groups, the WT mice were larger than the KO mice, as determined using DEXA. Body component mass assessment using DEXA (Fig. 1B) showed no differences between the WT-ND and KO-ND mice. In the HFD groups, the total mass was significantly lower in the KO group (39.87 ± 3.9 g) than in the WT group (46.78 ± 3.6 g). Similarly, the total fat mass was significantly lower (p < 0.01) in the KO group (13.09 ± 2.2 g) than in the WT group (19.46 ± 2.7 g). No differences were observed in lean mass, whereas fat mass in the tissues of the KO group was significantly lower than that in the WT group (p < 0.001). The eWAT weight of the KO-HFD group was significantly lower than that of the WT-HFD group, indicating that OMP KO alleviated HFD-induced obesity in mice without affecting lean mass (Fig. 1C).

OMP KO does not affect food intake and energy expenditure patterns
Knockout of the Omp gene is likely to affect the olfactory function, which may affect the hypothalamus, the brain region that regulates energy homeostasis. Therefore, we investigated whether the KO of the Omp gene affected the sense of smell or food intake. In the food-finding test, Omp-KO mice fed ND took longer to find food than those in the control group; however, when fed an HFD, food-finding time tended to be shorter, although not significanctly ( Fig. 2A). We also investigated whether OMP affected food intake. There was no difference in food intake between the two mouse genotypes in either the ND or HFD groups (WT-ND, n = 3; WT-HFD, n = 4; KO-ND, n = 4; KO-HFD, n = 4) (Fig. 2B).
To examine whether the lean phenotype observed in KO-HFD mice was related to changes in energy consumption, we measured these indicators in mice using metabolic cages over one week. O 2 uptake, CO 2 production, and the respiratory exchange ratio (RER) were similar between KO-HFD and WT-HFD mice (Fig. 2C-E). An RER value between 0.7 and 1.0 indicates that both fats and carbohydrates are utilized as the primary fuel sources. There was no difference in activity levels or heat production between the two groups ( Fig. 2F and G). There was no statistically significant difference in the metabolism levels between the two groups. Under normal dietary conditions, metabolic parameters were similar between the two mouse genotypes (Data not shown). Thus, the low body mass of Omp KO mice did not appear to be due to changes in energy consumption. These results suggest that Omp may affect adipocytes.

OMP deletion increases thermogenesis in BAT and decreases adipogenesis of WAT
After 13 weeks of diet-induced obesity, we examined the mRNA expression levels of Omp in WAT and BAT of the mice used in the experiment. Omp mRNA expression was not detected in the adipose tissue of the KO mice (Fig. S1A). H&E staining of BAT revealed that the WT-HFD mice had larger adipocytes than the KO-HFD mice (Fig. 3A). The lipid droplet percentage of BAT was significantly lower in the KO-HFD group (46.15 ± 10.53%) than that in the WT-HFD group (71.71 ± 10.53%) (p < 0.01) (Fig. 3A). The expression levels of thermogenesisand differentiation-related genes in the BAT of KO mice were higher compared to those in WT mice. Specifically, the expression of Ucp1, Ucp2, Vegfb and Pparα in BAT was significantly higher in OMP-KO mice than in WT mice under both feeding conditions. The C/EBPβ expression in BAT was not significantly different. However, a statistically significant difference in Fgf21 expression between WT and KO mice was observed only under HFD conditions (Fig. 3B). To investigate this further, we examined the morphology of the WAT. H&E staining of eWAT revealed that adipocytes in Omp-KO mice were smaller than those in WT mice under HFD conditions (Fig. 4A). Quantitative analysis indicated that adipocyte area of Omp-KO mice was smaller than that in the WT mice (p < 0.05) (Fig. 4A). In the case of eWAT, the mRNA expression levels of adipogenesis-related transcription factors, such as Pparγ, Srebf1, Cd36, Fasn, and Pparα, were lower in the KO-HFD group than those in the WT-HFD group (Fig. 4B).

OMP levels affect adipogenesis by regulating cAMP levels and CREB phosphorylation
We found that the protein expression levels of phosphorylated CREB were three-fold lower in Omp − /− MEFs than in Omp +/+ MEFs, and this decrease continued until 15 min after the induction of adipogenesis (Fig. 6A). There were no differences in cAMP levels between Omp +/+ and Omp − /− MEFs before adipogenesis induction. However, the cAMP levels in Omp − /− MEFs were 60% lower than those in Omp − /− MEFs after 5 min of differentiation. cAMP levels peaked at 15 min in both groups; moreover, cAMP levels were lower in Omp − /− MEFs than in Omp +/+ Heat production in 15-to 16-week-old WT and KO mice (males, n = 6) on an HFD were measured using the metabolic cage system. Values are shown as the mean ± SEM.
After treatment with forskolin, an AC activator, cAMP and CREB phosphorylation levels increased approximately 1.5 and 3.5-fold in Omp +/+ MEFs, respectively; however, a significant decrease was observed in Omp − /− MEFs ( Fig. 6C and D). Conversely, when the AC inhibitor SQ22536 was administered after two days of adipogenic differentiation induction with elevated cAMP levels, cAMP and CREB phosphorylation levels decreased significantly in Omp +/+ MEFs; however, no change was observed in Omp − /− MEFs (Fig, 6E and F). These data indicate that OMP affects CREB phosphorylation by modulating cAMP levels.
After transfection of mouse OMP into 3T3-L1 preadipocytes, we found that the levels of phosphorylated CREB increased 1.3-fold in Ompoverexpressing 3T3-L1 cells compared with that in the control (Fig. 6G). Moreover, Omp-overexpressing 3T3-L1 cells had higher cAMP levels than cells transfected with the empty vector (pcDNA3.1) at the same time points. This difference was significantly large (approximately 36%) 5 min after differentiation (Fig. 6H).

OMP deficiency leads to low adipogenesis through nuclear NF-κB activation
RNA sequencing of Omp +/+ MEFs and Omp − /− MEFs following treatment with forskolin revealed significant changes in the cAMP and NF-KB pathways (Data not shown). Specifically, mRNA expression levels of IκBα, responsible for inhibiting the NF-κB transcription, were 80% lower in Omp − /− MEFs than those in Omp +/+ MEFs (Fig. 7A). We subsequently measured the expression of IκBα and NF-κB using the samples shown in Fig. 6B. After treatment with forskolin, IκBα mRNA levels reduced and NF-κB levels increased in Omp − /− MEFs (Fig. S3A). As shown in Fig. 7B, after induction following the standard differentiation protocol for 6 h, nuclear NF-κB expression levels in Omp − /− MEFs were significantly higher than those in Omp +/+ MEFs. Furthermore, as shown in Fig. 7C, there was a clear trend of increased NF-κB activation in Omp − /− MEFs after induction following the standard differentiation protocol for 10 days. We also conducted experiments using Omp-overexpressing 3T3-L1 cells to determine the expression levels of IκBα and NF-κB following Omp-overexpression under adipocyte differentiation 8 per group), as assessed using RT-qPCR and normalized to that of Gapdh. Data are presented as the mean ± SEM. Statistical significance was calculated by Student's t-test for comparing two groups and two-way ANOVA with a Bonferroni's post-hoc test for multiple comparisons.**p < 0.01, ***p < 0.001 versus Omp +/+ ND group; #p < 0.05, ##p < 0.01 versus Omp +/+ HFD group.
conditions. However, the transfection efficiency decreased rapidly after 48 h, which complicated our experiments. As fully confluent cells requires 48 h before begin adipocyte differentiation induction, we were only able to analyze the early stage of differentiation, specifically the first hour after induction. The results revealed that the mRNA expression levels of IκBα increased in Omp-overexpressing 3T3-L1 cells compared with those in the control at 60 min. Furthermore, the mRNA expression levels of NF-κB decreased in Omp-overexpressing 3T3-L1 cells compared to those in the control at 15 and 60 min, but the difference was not statistically significant (Fig. S3B).

Discussion
Many studies on obesity prevention and adipogenesis have been conducted owing to the negative effects of obesity on health. Obesity and the olfactory system have been linked in many studies. According to a recent study, adult olfactory sensory neurons (OSNs) enhance thermogenesis in order to prevent diet-induced obesity (Riera et al., 2017). Research has also examined adipocytes and olfactory receptors (ORs) found in OSN cilia. Previous studies have revealed that OR 544 contributes to adiposity reduction , and mouse OR 23 is a regulator of adipogenesis and thermogenesis in 3T3-L1 cells (Tong et al., 2018). Here, we focused solely on OMP, a biomarker for the olfactory system, which is also expressed in non-olfactory systems (Kang and Koo, 2012).
In this study, we elucidated the signaling pathway through which Omp deficiency inhibits adipogenesis by regulating cAMP production and nuclear NF-κB activation. We confirmed that adipogenesis was suppressed in embryonic fibroblasts derived from Omp-KO mice. Moreover, we found that the gene expression and protein levels of PPARγ, a known master regulator of adipocyte differentiation (Lefterova et al., 2014;Siersbaek et al., 2010;Zhang et al., 2020), and that of the downstream target gene C/EBP were positively correlated with OMP expression levels during adipogenesis in 3T3-L1 cells. Furthermore, Omp deficiency suppressed the expression of transcription factors, such as Pparγ, Srebf1, Cd36, Fasn, and Pparα, that are involved in adipocyte differentiation in the WAT of HFD-fed mice.
Additionally, this study demonstrated that the absence of OMP diminished brown adipocyte dysfunction, which is commonly referred to as whitening, which compromises thermogenesis. We found that the (n = 8 per group), as assessed using RT-qPCR and normalized to that of Gapdh. Data are presented as the mean ± SEM. Statistical significance was calculated by Student's t-test for comparing two groups and two-way ANOVA with a Bonferroni's post-hoc test for multiple comparisons. #p < 0.05, ##p < 0.01 versus Omp +/+ HFD group.
BATs of OMP KO-HFD mice had lower lipid droplet area than those of WT-HFD mice. Furthermore, mRNA expression analyses indicated a significant upregulation of thermogenesis-related genes, including Ucp1, Ucp2, Vegfb, and Pparα in the BAT of OMP KO-HFD mice compared to those in WT-HFD mice (Ziqubu et al., 2023). For metabolic studies, body composition measurements were performed by weighing dissected tissues ex vivo from euthanized animals. Determination of body composition in vivo using DEXA simplifies the procedures and has been successfully applied to mice (Windahl et al., 1999). The study conducted by Sjögren et al. showed that the fat percentage data derived from DEXA correlated with the amount of adipose tissue dissected in mice (Sjogren et al., 2001). However, DEXA results merely provide a 0.9142, p < 0.0001), C/EBPα (R 2 = 0.8926, p = 0.0001), and C/EBPβ (R 2 = 0.9519, p < 0.0001) during adipogenesis in 3T3-L1 preadipocytes at 0, 1, 2, 3, 4, 5, 6, 7, and 8 days, as determined using western blotting. (C-D) Relative protein expression of key transcription factors involved in adipogenesis was assessed using western blotting, and triangle pointed bands (C/ EBPβ, C/EBPα and PPARγ) were quantified and normalized to that of GAPDH. Levels of protein are expressed as fold value Omp +/+ MEFs at 0 min. Each point represents an independent replicate of the western blot experiment. Data are presented as the mean ± SEM. Statistical significance was calculated using Student's ttest. *p < 0.05 versus Omp +/+ MEFs at the same time point.
general estimate and must be contrasted with anatomical dissection and weighing to provide an accurate estimation of fat percentage. In our study, DEXA body component mass revealed that in HFD-fed mice, the fat mass in the tissues of the KO group was significantly lower than that in the WT group. Omp-deficient mice had low-fat mass, which is normally associated with small adipocytes, and are resistant to HFD-induced obesity. However, there are some potential adverse effects of a whole-body OMP KO, such as olfactory dysfunction, including a loss of the sense of smell, compared to an adipocyte-specific transgenic mouse model. The data obtained from this study showed that OMP did pcDNA3.1 vector for 48 h in 3T3-L1 preadipocytes, cells were subjected to a standard differentiation protocol for 15 min, followed by the measurement of CREB phosphorylation and cAMP levels. Levels of cAMP are expressed as -fold empty pcDNA3.1 vector-transfected 3T3-L1 cell at 0 min *p < 0.05, **p < 0.01 versus the pcDNA3.1-transfected control. p-CREB: phosphorylated CREB; DMSO: dimethyl sulfoxide; FSK: forskolin; Data are presented as the mean ± SEM. Statistical significance was calculated by Student's t-test for comparing two groups and two-way ANOVA with a Bonferroni's post-hoc test for multiple comparisons. ns, not significant. *p < 0.05, **p < 0.01, ***p < 0.001 versus Omp +/+ MEFs at the same time point. (C) NF-κB p65 levels during adipogenesis induced following the standard differentiation protocol in Omp +/+ MEFs and Omp − /− MEFs for 10 days and normalized to those of GAPDH (Cytoplasm protein) or LAMIN B1 (Nuclear Protein). NF-κB: nuclear factor kappa B; IκBα: inhibitor of κB; FSK: forskolin; Data are presented as the mean ± SEM. Statistical significance was calculated by Student's t-test for comparing two groups and two-way ANOVA with a Bonferroni's post-hoc test for multiple comparisons. **p < 0.01, ***p < 0.001 versus Omp +/+ MEFs. not affect food intake or food-finding time in the HFD group. This is consistent with the findings of Riera et al. OMP did not affect food intake (Riera et al., 2017).
We used only male mice, as they are more susceptible to developing obesity and associated metabolic disorders (Maric et al., 2022). Using only male mice allowed us to reduce the potential confounding effects of hormonal fluctuations associated with the estrous cycle in female mice. However, future studies using both male and female mice as well as other species could provide a more comprehensive understanding of the effects of OMP on metabolic health. Furthermore, our findings support the hypothesis that OMP plays a role in adipogenesis by regulating cAMP levels. Several studies have shown that cAMP is a fundamental messenger involved in olfactory transduction (Fleischer et al., 2009;Kwon et al., 2009). Additionally, OMP serves as a major cAMP regulator that maintains ORN sensitivity and plays a critical role in various biological processes (McKnight et al., 1998). During adipogenesis, preadipocytes that are activated by substances that increase intracellular cAMP levels differentiate into mature adipocytes following the activation of adipogenesis-related gene expression (Tang and Lane, 2012).
In our study, Omp-overexpressing 3T3-L1 cells showed high cAMP levels at the early stages of adipogenesis, whereas the Omp-deleted group showed low cAMP levels overall. A more compelling finding was that OMP-deficient cells showed almost no response to an AC activator or inhibitor after the induction of differentiation.
Previous studies on adipocyte differentiation have shown that CREB is a major target for cAMP signalling and that CREB activation promotes adipocyte differentiation by activating transcription factors, which in turn regulate gene expression that confers the adipocyte phenotype (Tang and Lane, 2012;Fox et al., 2008). In our study, Omp-overexpressing 3T3-L1 cells showed high levels of CREB phosphorylation after adipogenic induction, whereas Omp − /− MEFs showed low levels of CREB phosphorylation. These data suggest that the absence of OMP affected CREB phosphorylation by decreasing cAMP levels, thereby reducing adipogenesis.
The results of the above studies led us to identify and investigate the pathways through which the reduction in adipocyte differentiation occurs. In our study, RNA sequencing analysis revealed that the expression of IκBα following forskolin treatment was low in Omp − /− MEFs. Based on these results, the activation of NF-κB signalling by Omp deficiency provides a key clue for elucidating the effects of OMP on adipocyte differentiation.
While NF-κB signaling activation inhibits adipogenesis, the differentiation of preadipocyte 3T3-L1 cells is recovered upon its inhibition (Chae and Kwak, 2003). The role of cAMP in NF-κB signaling appears to be highly dependent on the cell type and circumstances. For example, cAMP inhibits NF-κB activity in 3T3-L1 fibroblasts (Dobashi et al., 2003) but stimulates it in breast cancer cells . Many studies have demonstrated the inhibitory effects of cAMP on NF-κB function caused by IκB activity, which inhibits the translocation of NF-κB from the cytoplasm to the nucleus (Kamthong and Wu, 2001). In our study, high expression levels of the nuclear NF-κB subunit p65 and a significant decrease in IκB-α activity during differentiation of Omp − /− MEFs suggest that NF-κB signaling is activated at a high level when OMP is absent.

Conclusion
Our findings indicate that OMP has a significant biological impact beyond its effects on the olfactory response. The results revealed that OMP is responsible for regulating cAMP kinetics during adipogenesis and affects its downstream mechanisms. OMP deficiency inhibits adipogenesis by increasing NF-κB nuclear translocation via decreasing cAMP and phospho-CREB protein expression during adipogenesis. Finally, our results suggest that ectopic olfactory signalling events during adipogenesis may provide new perspectives for obesity research.

Declaration of competing interest
None.

Data availability
Data will be made available on request.